S. A. Yemini, G. S. Goldszmidt, A. D. Stoyenko, Y.-H. Wei and L. Beeck, `CONCERT: a high level language approach to heterogeneous distributed systems', proc. 9th IEEE International Conference on Distributed Computing Systems, Newport Beach, CA, 5--9 June, 1989.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....debuggers and a means of replaying pro56 gram executions so that POET could be used to control a debugging session. 5.1. 2 Architecture POET is target system independent [27] While originally used with the Hermes environment [24] it has since been used with ABC [2] C [5] Concert C [28], OSF DCE [11] Java [15] PVM [14] SR [1] and to debug itself. Configuration information for each environment is stored in a target description file external to the program. The relevant environment is identified when the first target program is started and POET configures itself appropriately. ....

Shaula A. Yemini, German S. Goldszmidt, Alexander D. Stoyenko, Yi-Hsiu Wei, and William Beeck. CONCERT: A high-level-language approach to heterogeneous distributed systems. In Proceedings of the 9th International Con- ference on Distributed Systems, Los Alamitos, CA., 1989. IEEE Computer Society Press.

....abstractions are close to the hardware and implementations typically need complete control. There are systems that offer abstractions close to application programming constructs. Amber s [Ch89] passive and active objects allow the programmer to develop flexible object oriented programs. Concert [Ye89] extends standard programming languages such as C and PL 1 to implement heterogenous cooperative peer processing. Such systems are typically implemented on top of an existing operating system, Amber on Topaz and Concert on OS 2 and VM 370. However, they confine the programmer to a particular ....

Yemini, S., et. al., CONCERT: A High-Level-Language Approach to Heterogeneous Distributed Systems, Proceedings of the Ninth International Conference on Distributed Computing, June

....this goal can be approximated conservatively by using an interconvertibility type system and an extensible set of canonicalizing transformations. 5. We showed how code is automatically generated from MockSL. The Mockingbird project builds on the insights and experiences of the Concert project [18, 3, 4, 5], particularly the idea of using a specialized internal representation for interfaces [5] However, in Mockingbird we have organized the compilation system more explicitly around the problem of interoperation. In addition, the Mockingbird tools are implemented so as to avoid either nonstandard ....

S. A. Yemini, G. Goldszmidt, A. Stoyenko, Y. Wei, and L. Beeck. Concert: A high-levellanguage approach to heterogeneous distributed systems. In The Ninth International Conference on Distributed Computing Systems, pages 162--171. IEEE Computer Society, June 1989. 16

....familiar with the language. However, such languages are, in general, poor at accommodating legacy code, and require new programmer skills. At the opposite extreme, commercial packages (such as OSF DCE [26] Apollo NCS [19] or SUN RPC [31] and software tools (such as Matchmaker [18] Courier [33], Horus [16] and HRPC [9] do a good job of accommodating legacies, and sometimes heterogeneity. The stub compilers associated with these systems also hide some of complexity associated with marshalling data. However, as we discuss in [6] library based tools do a poor overall job of hiding ....

S. A. Yemini, G. Goldszmidt, A. Stoyenko, Y. Wei, and L. Beeck. Concert: A high-level-language approach to heterogeneous distributed systems. In The Ninth International Conference on Distributed Computing Systems, pages 162--171. IEEE Computer Society, June 1989.

....an ensemble. In Hermes, this coordination model is its process model. Hermes processes are created and terminated dynamically and communicate and synchronize by synchronous and asynchronous message passing. Not only has this model been incorporated into a number of other computational languages [96], the embodied messagepassing paradigm is a very common one in distributed computing. This facilitates porting our results to environments with a similar coordination model. Second, processes in Hermes are both the unit of parallelism and modularization, see [34] Even a moderate Hermes ....

Shaula A. Yemini, German S. Goldszmidt, Alexander D. Stoyenko, and Yi-Hsiu Wei. CONCERT: A High--Level--Language Approach to Heterogeneous Distributed Systems. In Proceedings of the 9th International Conference on Distributed Computing Systems, pages 162--171, Newport Beach, California, USA, June 1989.

....[12] However, the modules of the system must still be concerned with when to send or receive messages on these ports. In Polylith [16, 17] a configuration is expressed using module interconnection constructs that establish procedure call bindings among modules in a distributed system. CONCERT [21] provides a uniform communication abstraction by extending several procedural programming languages to support the Hermes [19] distributed process model. PROFIT [9] provides a mixture of data sharing and RPC communication through facets with data and procedure slots that are bound to slots in ....

Shaula A. Yemini, German S. Goldszmidt, Alexander D. Stoyenko, and Langdon W. Beeck. CONCERT: A high-level-language approach to heterogeneous distributed systems. In Proceed19 ings of the 9th International Conference on Distributed Computing Systems, pages 162--171, 1989.

.... owner (i.e. the user running the program) who delivered the resource to the resource manager, and the set of users who are to be given access the resource. The Hermes project also supports communication and interoperability with other languages and machine architectures. The Concert C project [37] supplies an interface definition language for specifying contracts [38] to which communicating processes adhere. A contract defines the promises made by the sender and the assumptions of the receiver. Using contracts, Concert processes may communicate securely with Hermes processes. 2.4 ....

S. A. Yemini et al.. Concert: A high-level language approach to heterogenous distributed systems. In The Ninth International Conference on Distributed Computing Systems, pages 162--171. IEEE Computer Society, June 1989.

....terms of connections between communication ports, the Polylith view of a configuration is expressed in terms of a set of procedure call bindings. Polylith provides a set of module interconnection constructs for establishing procedure call bindings among modules in a distributed system. CONCERT [23] provides a coordination language in which the Hermes [22] distributed process model is embedded in several procedural programming languages by providing extensions to each language. 1 A more extensive discussion on the benefits of the separation of structure and function is given elsewhere ....

Shaula A. Yemini, German S. Goldszmidt, Alexander D. Stoyenko, and Langdon W. Beeck. CONCERT: A high-level-language approach to heterogeneous distributed systems. In Proceedings of the 9th International Conference on Distributed Computing Systems, pages 162--171, 1989.

....at odds with our high performance objectives. Therefore, we have decided to sacrifice upward compatibility. 2. 2 Approach We consider a metasystem to be an combination of two different types of systems, parallel processing systems (PPS) and heterogeneous distributed computing systems (HDCS) [1,2,4 6,12,20,21,24,25,29,31]. Borrowing from the object oriented lexicon, any solution to the metasystems problem will inherit attributes and behaviors from both areas (Figure 2 below) While it is important that we inherit many features from both PPS and HDCS, some features of these systems are at odds with one another. For ....

....Speedup relative to IPC time Speedup relative to SGI time IPC sequential time (seconds) SGI sequential time (seconds) 256, 256) 173 3.87 2.1 672 368 (341, 171) 121 5.53 3.03 672 23 4. 0 Related work Heterogeneous distributed computing systems have been an active area of research for some time [1,2,46, 12,20,21,24,25,29,31]. Our work differs from the work in the HDCS community in our objectives and the trade offs we are willing to make. The primary objectives in much of the HDCS work are interoperability, sharing, and availability. Unlike our work, high performance is not the objective. Applications portability ....

S. A. Yemini, et al., "Concert: A High-Level-Language Approach to Heterogeneous Distributed Systems", Proc. 9th IEEE International Conference on Distributed Computing Systems, Newport, CA., pp. 162-171, June,

....[12] However, the modules of the system must still be concerned with when to send or receive messages on these ports. In Polylith [15] a configuration is expressed using module interconnection constructs that establish procedure call bindings among modules in a distributed system. CONCERT [19] provides a uniform communication abstraction by extending several procedural programming languages to support the Hermes [17] distributed process model. PROFIT [10] provides a mixture of data sharing and RPC communication through facets with data and procedure slots that are bound to slots in ....

Shaula A. Yemini, German S. Goldszmidt, Alexander D. Stoyenko, and Langdon W. Beeck. CONCERT: A highlevel -language approach to heterogeneous distributed systems. In Proceedings of the 9th International Conference on Distributed Computing Systems, pages 162--171, 1989.

....conclusions and suggestions for further work, including extension of the process clustering concept to events. The prototype software which is described in Section 3 was originally developed for the Hermes environment. Although it has now been retargeted to several other environments (Concert C [3], OSF DCE [6] SR [1] and the System [4] the description in the paper is oriented to the Hermes environment. That environment was used for initial development of the hierarchical clustering facilities because the large number of processes in a Hermes application presents significant ....

S. Yemini, et al, "CONCERT: A high-levellanguage approach to heterogeneous distributed systems," Proceedings of the 9th International Conference on Distributed Computing Systems, pp. 162-171 (June 5-9, 1989).

....re execution of the application, coercing it to follow the same partial order as during the initial execution. The tool was used initially with programs written in Hermes [7] but use with other environments was contemplated from the beginning and extension to other environments, such as Concert C [10] and OSF DCE [6] began once the prototype software became reasonably stable. As the software was modified, continuing efforts were taken to retain and enhance its target system independence. The tool is known as POET (for PartialOrder Event Tracer ) and has a multi process structure, as ....

S. Yemini et al. Concert: A high-levellanguage approach to heterogeneous distributed systems. In Proceedings of the 9th International Conference on Distributed Computing Systems, pages 162--171, June 5--9 1989.

....are expressed in a separate computation language that allows ports to be declared for interconnection within Darwin. In Polylith [27, 28] a configuration is expressed using module interconnection constructs that establish procedure call bindings among modules in a distributed system. CONCERT [36] provides a uniform communication abstraction by extending several procedural programming languages to support the Hermes [32] distributed process model. PROFIT [15] provides a mixture of data sharing and RPC communication through facets with data and procedure slots that are bound to slots in ....

Shaula A. Yemini, German S. Goldszmidt, Alexander D. Stoyenko, and Langdon W. Beeck. CONCERT: A high-level-language approach to heterogeneous distributed systems. In Proceedings of the 9th International Conference on Distributed Computing Systems, pages 162--171, 1989.

....with support for shared memory communication and integrated performance monitoring. It promotes the exploitation of heterogeneity, as opposed to attempting to hide it. The design of HBD has also been influenced by a number of other systems; e.g. Agora [5] Marionette [13] HIGHLAND [15] Concert [20], PHARROS [8] and Mentat [9] 4.1 Motivational Analogy To motivate the design of the HBD programming environment, let us make an analogy to human systems. Many of the most difficult and important open research problems cross disciplines (e.g. medicine, biology, and computer science) In order to ....

S. A. Yemini, C. S. Goldsmidt, A. D. Stoyenko, Y.-H. Wei, and L. W. Beeck. CONCERT: A High-Level Language Approach to Heterogeneous Distributed Systems. In Proc. 9th International Conference on Distributed Computing Systems, IEEE, June 1989, pp. 162-171.

....part of this code is a table that defines important characteristics of the events in the target environment, such as what event types pair with each other and what symbol should be used to display an event. The prototype has now been adapted to work with four other target environments: Concert C [21], SR [1] the System [7] and OSF DCE [17] The differences among these environments are substantial, which suggests that the prototype can indeed be adapted to a wide variety of target environments by changing only a very small part of the code. In particular, although the above description ....

S. Yemini et al. CONCERT: A high-level-language approach to heterogeneous distributed systems. In Proc. 9th Int. Conf. on Distributed Computing Systems, pages 162--171, June 5--9 1989.

....a distributed and parallel programming resource appropriate for a homogeneous set of hosts. The HCS project [NoBL88] shows one way for providing a heterogeneous RPC capability in a distributed environment; more recently, a lightweight remote procedure call was demonstrated [BALL90] Concert [YGSW89] and Marionette [SuAn89] are more variations on a theme. Several early projects emphasized a network filesystem approach (such as Locus [PWCE81] An interesting approach to cross architecture procedure call using a common backing store is given by Essick [Essi87] The ISO OSI framework for ....

Yemini, S., G. Goldszmidt, et al. CONCERT: A high-level language approach to heterogeneous distributed systems, Proc of 9th Conf on Distributed Computing Systems, (June 1989), pp. 162-171.

....and the database systems available. These differences must be masked. 2. 1 Approach We consider a metasystem to be an combination of two different types of systems, parallel processing systems (PPS) and heterogeneous distributed computing systems (HDCS) 1] 2] 4] 5] 6] 11] 13] 22] 23] 25] 26] 30][31][21] Borrowing from the object oriented lexicon, any solution to the metasystems problem will inherit attributes and behaviors from both areas (Figure 2) While it is important that we inherit many features from both PPS and HDCS, some features of these systems are at odds with one another. For ....

....time SGI sequential time (seconds) IPC sequential time (seconds) 256, 256) 173 3.87 2.1 368 672 (341, 171) 121 5.53 3.03 368 22 ble alternative. 4. 0 Related work Heterogeneous distributed computing systems have been an active area of research for some time [2] 4] 5] 6] 13] 22] 23] 25] 26] 30][31]. Our work differs from the work in the HDCS community in our objectives and the trade offs we are willing to make. The primary objectives in much of the HDCS work are interoperability, sharing, and availability. Unlike our work, high performance is not the objective. Applications portability ....

S. A. Yemini, et al., "Concert: A High-Level-Language Approach to Heterogeneous Distributed Systems", Proc. 9th IEEE International Conference on Distributed Computing Systems, Newport, CA., pp. 162-171, June,

....an ensemble. In Hermes, this coordination model is its process model. Hermes processes are created and terminated dynamically and communicate and synchronize by synchronous and asynchronous message passing. Not only has this model been incorporated into a number of other computational languages [67], the embodied message passing paradigm is a very common one in distributed computing. This facilitates porting our results to environments with a similar coordination model. Second, processes in Hermes are both the unit of parallelism and modularization, see [22] Even a moderate Hermes ....

Shaula A. Yemini, German S. Goldszmidt, Alexander D. Stoyenko, and Yi-Hsiu Wei. CONCERT: A High--Level--Language Approach to Heterogeneous Distributed Systems. In Proceedings of the 9th International Conference on Distributed Computing Systems, pages 162--171, Newport Beach, California, USA, June 1989.

No context found.

S. A. Yemini, G. S. Goldszmidt, A. D. Stoyenko, Y.-H. Wei and L. Beeck, `CONCERT: a high level language approach to heterogeneous distributed systems', proc. 9th IEEE International Conference on Distributed Computing Systems, Newport Beach, CA, 5--9 June, 1989.

....instructions, far less than in any lightweight process system. NIL was evaluated by using it to prototype two large SNA subsystems. This experience was encouraging both from the standpoint of software engineering and from the standpoint of performance. 5. 2 Concert The existing Concert prototypes [15] demonstrate processes written in C running under AIX or OS 2 on PS 2 hardware calling each other and also VM 370 processes written in PL I. Supported communications protocols include 3270 terminal emulation, APPC, NETBIOS, and TCP IP. To bridge the heterogeneity of these environments, a ....

S. A. Yemini, G. Goldszmidt, A. Stoyenko, Y. Wei, and L. Beeck, "Concert: A high-levellanguage approach to heterogeneous distributed systems," in The Ninth International Conference on Distributed Computing Systems, pp. 162--171, IEEE Computer Society, June 1989.

....languages. Thus, we may build Concert Cobol, Concert Fortran, Concert C , etc. Each of these languages will implement a distributed computing model called the process model: a distributed program is composed of a set of sequential processes communicating by RPC and asynchronous message passing [35, 2]. These languages will interoperate, so that a Concert Cobol process could serve an RPC by a Concert C process. To support data interoperability, IPC messages are mapped in and out of a Concert Universal type family, as described in [6] We are expanding the support for OSF DCE to further exploit ....

S. A. Yemini, G. Goldszmidt, A. Stoyenko, Y. Wei, and L. Beeck. Concert: A high-level-language approach to heterogeneous distributed systems. In The Ninth International Conference on Distributed Computing Systems, pages 162--171. IEEE Computer Society, June 1989.

....add to and to reduce the functionality of a server during execution time. It enables a more flexible distribution of functionality between processes, for example, a process can dynamically transfer services to another process. This scheme builds on a multiapplication, multilanguage process model [12, 2] that supports instantiation, interconnection, and communication of processes. Client processes can delegate to elastic servers the execution of delegated programs. Elastic servers instantiate delegated programs when necessary, and as determined locally by the server or via a client request. ....

Yemini S., Goldszmidt G., Stoyenko A., Wei Y., and Beeck L. Concert: A highlevel -language approach to heterogeneous distributed systems. In The Ninth International Conference on Distributed Computing Systems, pages 162--171. IEEE Computer Society, June 1989.

No context found.

S. Yemini et al. CONCERT: A high-level-language approach to heterogeneous distributed systems. In Proc. 9th Int. Conf. on Distr. Comp. Systems, pages 162--171, June 1989.

No context found.

S.A. Yemini, G.S. Goldszmidt, A.D. Stoyenko, Y. Wei, and L. Beeck. Concert: A high-level-language approach to heterogeneous distributed systems. In Proceedings of the 9th International Conference on Distributed Computing Systems, pages 162--171, June 1989.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC